Multitask System Design by an Improved PSO Algorithm

Abstract

This paper will propose a simple modified version of the particle swarm optimization (PSO) algorithm to solve multitasking problems. In the proposed scheme, a single population existed in the original PSO algorithm is firstly divided into several subpopulations and the number of subpopulations entirely depends on the number of the problems that will be solved. The related information regarding the best particle within each subpopulation needs to be recorded, and each of them is to deal with one corresponding problem. To show the applicability and efficiency of the developed method, some target searching problems with two dimensions are examined.